Posudky závěrečné kvalifikační práce

Posudek vedoucího

Rougemaille, Nicolas

First of all, I would like to emphasize that the diploma thesis was entirely prepared at CEITEC, while I am working at the Néel Institute in Grenoble, France. All interactions and discussions I had with Vojtěch Schánilec were done by skype. I thus followed his work from abroad and he conducted everything on his own (sample fabrication and characterization, magnetic imaging, development of the software for the image analysis, setting up a demagnetization stage, running micromagnetic simulations). As a matter of fact, he did an ERASMUS internship last year with us and he acquired a background on the topic during his 6 month stay in Grenoble (Feb 2017 – July 2017). Although he did not start from zero, during his diploma thesis he studied a different artificial spin ice system, having a more complex physics than the square ice system he investigated during his ERASMUS project. I was impressed by the work he performed at CEITEC and the quality of the results he got, both regarding the sample fabrication and the magnetic imaging. In addition, although I was not very enthusiastic when he decided to develop a code for analyzing his magnetic images, I was again impressed by the work he performed and the quality of the output. He also took the lead to run micromagnetic simulations on his own, both to compute vertex energies and to explain the magnetic contrast obtained in magnetic force microscopy measurements. To summarize, he handled all aspects of the project, from the sample design to its characterization / simulation and the data analysis. For this reason, I feel fair to evaluate his experimental skills by a A mark. Besides, Vojtěch Schánilec is a well-organized, hard working person. I enjoyed very much working with him and his work will be published in several peer-reviewed journals. In fact, our collaboration was so efficient and pleasant that we decided to continue working together in the framework of a PhD thesis for which he successfully got a grant (Brno / Grenoble cotutelle starting this fall). Below I provide a few questions you may want to ask during the defense. We never talked together about those questions. They might be tricky and could be used to challenge him a bit (I do not have clear answers myself to those questions and they might be used to further discuss the results). 1) Do you understand why the edges of the array behave differently than the bulk? Is this an intrinsic property of a finite size system, or is this related to the fabrication process? Personal feeling: I wonder whether the lithography process can affect the magnetic properties of the array, for example because the electron dose during the fabrication process is different at the edges than in the bulk. 2) When demagnetizing your sample, you stabilize crystallites of the ground state configuration (LRO), and by changing the demagnetization time, you change the average size of those crystallites. Why magnetic configurations with small LRO crystallites resemble a spin ice 1 state if these configurations are built from LRO crystallites? Personal feeling: At some point, the LRO crystallites are so small that we do not see any difference with a conventional spin ice 1 state. Or otherwise said, there are so many domain walls between small LRO crystallites that domain walls contribute more than LRO crystallites. 3) Do you think the physics would work the same if you were able to make your system thermally active, like in Chioar's paper where a GdCo alloy with a low Curie temperature was used? Personal feeling: That should work the same as the kinetic algorithm described in Chioar's paper leads to the same physics as the one provided by Monte Carlo simulations. 4) Do you think your ''notch'' approach could be used for other spin ice geometries? No personal feeling... maybe in systems having an ordered ground state that is hard to reach because of the existence of disordered manifolds at higher temperatures.

Dílčí hodnocení
Kritérium	Známka	Body	Slovní hodnocení
Splnění požadavků a cílů zadání	A
Postup a rozsah řešení, adekvátnost použitých metod	A
Vlastní přínos a originalita	A
Schopnost interpretovat dosažené výsledky a vyvozovat z nich závěry	B
Využitelnost výsledků v praxi nebo teorii	A
Logické uspořádání práce a formální náležitosti	A
Grafická, stylistická úprava a pravopis	B
Práce s literaturou včetně citací	B
Samostatnost studenta při zpracování tématu	A

Navrhovaná známka: A

Posudek oponenta

Hrabec, Aleš

The work of Bc. Vojtech Schanilec deals with the development and description of artificial spin ice, a mesoscopic system which is used to mimic properties of frustrated systems occurring in nature. Here, a modified kagome lattice is used to achieve different states of the spin ice. The student demonstrated a number of complex skills during this process: from designing a smart trick to facilitate the path towards low energy spin ice states, micromagnetic simulations, lithography, growth of the magnetic films, development of MFM tips, MFM measurements and, most remarkably, he developed a code for the analysis of MFM images. This is particularly valuable because such software allows avoiding the manual treatment of the MFM contrast and opens a path for future studies of large arrays of nanomagnets of different geometries. The main drawback of this work, which is caused by the particular health circumstances, is the manuscript. Despite attempts to guide a reader through the text, it is difficult to follow the text and one needs to consult the (correctly placed) cited literature and I believe that it will be difficult for new students to use the manuscript as a starting point. The manuscript contains a number of typos and errors and its corrected version was sent separately. Despite this fact, from my point of view, the work has a very high scientific value and quality and will have with no doubt an impact on international community of statistical physics. This is best illustrated by the fact that the work presented here is also submitted into a peer reviewed journal. Based on this, I can strongly support and recommend this work to be presented at the final exam.

Dílčí hodnocení
Kritérium	Známka	Body	Slovní hodnocení
Splnění požadavků a cílů zadání	A
Postup a rozsah řešení, adekvátnost použitých metod	A
Vlastní přínos a originalita	A
Schopnost interpretovat dosaž. výsledky a vyvozovat z nich závěry	B
Využitelnost výsledků v praxi nebo teorii	A
Logické uspořádání práce a formální náležitosti	B
Grafická, stylistická úprava a pravopis	C
Práce s literaturou včetně citací	A

Navrhovaná známka: A

Otázky

1) The entire field of spin ice is based on the fact that finite entropy can be present even at zero temperature which violates the third law of thermodynamics. The most typical example of such a system is water, as described in the thesis and illustrated in Fig.2.6. Can you simply explain what is microscopically happening in water at zero temperature, i.e. how the system physically manifests the non-zero entropy?
2) The experimental results are throughout the work compared to the Monte-Carlo simulations published by Chioar et al. These calculations were performed for a kagome lattice containing islands with dimensions of 500x100x10 nm3 fabricated from a GdCo alloy. However, in the presented experimental work, Permalloy islands with dimensions of 1000x250x25 nm3 were used. Therefore apart from the micromagnetic parameters (magnetization and exchange) also the geometry is different. Is it still justifiable to compare the experimental results with the particular theoretical curves? Can you also explain the difference between positive and negative correlation coeffieicents?
3) The experimental concept of artificial spin ice, developed by Wang and colleagues, is based on the fact that each island can be represented by a macrospin (Ising model). Such spins then interact via dipolar interaction described by equation 2.8. However, the magnetic islands used here are physically connected and interact not only via dipolar interaction but also by the exchange coupling. What parameter or fact justifies that this system still behaves as a spin ice?
4) Following the previous question: It is concluded in Chapter 6 that bigger defects (notches) force the system into the lower effective temperature state. This gives me a feeling that the system with the largest notch would be the one with completely disconnected islands. Have you you performed similar measurements on a classical kagome lattice, i.e. containing individual (disconnected) islands?
5) On p.14 is stated that ASI systems are an athermal system, and therefore flipping of the macro spins is impossible by temperature. This is in a strong contradiction with the rest of the work where it is discussed how such a system can be effectively thermalized. Can you in a few words explain the difference between superparamagnetic islands and islands heated up above Curie temperature? Which parameters set the behavior of superparamagnetic islands, i.e. the energy barrier of the spin to flip from one direction into another?
6) In the work, it is also mentioned that the resolution of MFM is given by the height of fly during the second scan when the information about the magnetic configuration is recorded. The distance used in this work is in the range 40-90nm. Why would one use larger distances to perform such scans if the spatial resolution becomes worse? What experimental technique would you use to reveal the absolute orientation of the spins within the islands?
7) This is just a final remark. In Fig.5.2 is shown how the edge states perturb the ideal behaviour of infinite spin ice. It would be interesting to see how the size of the system affects the residual entropy, i.e. when the dimensionality of the problem is reduced.